The NERC and CEH trademarks and logos ('the Trademarks') are registered trademarks of NERC in the UK and other countries, and may not be used without the prior written consent of the Trademark owner.

Abstract1. Below-ground resource partitioning is often proposed as the underlying mechanism for the positive relationship between plant species richness and productivity.For example, if species have different root distributions, a mixture of plant species may be able to use the available resources more completely than the individual species in a monoculture. However, there is little experimental evidence for differentiation in vertical root distributions among species and its contribution to biodiversity effects.2. We determined species-specific root standing biomass over depth using molecular techniques (real-time qPCR) in a large grassland biodiversity experiment (one to eight plant species mixtures), in 2 years. Species-specific root biomass data were used to disentangle the effects of positive interactions between species (complementarity effects) and effects due to dominance of productive species (selection effects) on root biomass in mixtures. In a next step, these biodiversity effects were linked to the diversity of rooting depths and the averaged rooting depth of the community.3. Root biomass increased with species richness. This was mainly due to positive interactions (the complementarity effect), which increased with species richness below-ground. In contrast, the selection effect decreased with species richness.Although there was considerable variation in vertical root distribution between species in monocultures, the diversity of rooting strategies did not explain the complementarity effect. Rather, the abundance of deep-rooting species in mixtures (i.e. high community-weighted mean) was significantly related to the complementarity effect. Comparing the "predicted" root distribution (based on monocultures) to the actual distribution in mixtures, we found that mixtures rooted deeper than expected, but this did not better explain the complementarity effect.

Soil amendment with biochar is currently proposed as a management strategy to improve soil quality and enhance plant productivity. Relatively little is known about how biochar affects plant competition, although it has been suggested that it can increase the competitive ability of legumes. This study tested the impact of a biochar on the competitive ability of legumes through alterations to soil pH and/or nutrient availability. Biochar was produced from aboveground plant biomass from a species-rich semi-natural grassland pyrolysed at 400°C. In a greenhouse experiment, a legume (red clover, Trifolium pratense L.); a grass (red fescue, Festuca rubra L.); and a forb (plantain, Plantago lanceolata L.) were grown in (1) monocultures, (2) in a mixed culture of red fescue and red clover, and (3) in a mixture of all three species. Soil treatments included fertilization with nitrogen (N), potassium (K), phosphorus (P), or micronutrient fertilizer in the presence or absence of biochar; a pHadjusted control soil; and a control (i.e. with no amendment). The competitive ability of red clover was quantified as the proportion of aboveground biomass of this species within the mixtures. Both biochar amendment and K fertilization significantly (P < 0.001) increased red clover biomass, and increased the competitive ability of red clover when grown with red fescue and plantain. Application of N fertilizer, irrespective of biochar amendment, resulted in significantly (P < 0.001) greater red fescue and plantain biomass and eliminated the competitive advantage of red clover. The biochar-mediated pH increase did not affect red clover biomass or its competitive ability. We conclude that biochar has a beneficial effect on red clover under N limiting conditions due to an increase in K availability. Our results suggest a potential role for biochar to maintain the proportion of forage legumes in agricultural pastures or semi-natural grasslands.

Increased biological nitrogen fixation (BNF) by legumes has been reported following biochar application to soils, but the mechanisms behind this phenomenon remain poorly elucidated.We investigated the effects of different biochar application rates on BNF in red clover (Trifolium pratense L.) Red clover was grown in mono or mixed cultures with red fescue grass (Festuca rubra L.) and plantain (Plantago lanceolata L.) at a range of different biochar application rates (0, 10, 50 and 120 t ha -1 ). In a separate experiment, nutrient effects of biochar on BNF were investigated using nitrogen, phosphorous and potassium (N, P, K) and micronutrient fertilisation using the same plant species.Biochar addition increased BNF and biochar applied at a rate of 10 t ha -1 led to the highest rate of BNF. Total biomass also showed the greatest increase at this application rate. An application rate of 120 t ha -1 significantly decreased biomass production in both single and mixed cultures when compared to the control, with the greatest reduction occurring in red clover. Furthermore, BNF was significantly higher in pots in which red clover was grown in mixed cultures compared to monocultures. In the absence of biochar, K fertilization caused a significant increase in BNF. For N, P, and micronutrient fertilization, BNF did not significantly differ between treatments with and without biochar addition.We conclude that different biochar applications rates lead to different effects in terms of BNF and biomass production. However, due to the high variety of biochar properties, different application rates should be investigated on a case specific basis to determine the optimum biochar application strategies.4

Plant diversity influences many ecosystem functions including root decomposition. However, due to the presence of multiple pathways via which plant diversity may affect root decomposition, our mechanistic understanding of their relationships is limited. In a grassland biodiversity experiment, we simultaneously assessed the effects of three pathways-root litter quality, soil biota, and soil abiotic conditions-on the relationships between plant diversity (in terms of species richness and the presence/absence of grasses and legumes) and root decomposition using structural equation modeling. Our final structural equation model explained 70% of the variation in root mass loss. However, different measures of plant diversity included in our model operated via different pathways to alter root mass loss. Plant species richness had a negative effect on root mass loss. This was partially due to increased Oribatida abundance, but was weakened by enhanced root potassium (K) concentration in more diverse mixtures. Equally, grass presence negatively affected root mass loss. This effect of grasses was mostly mediated via increased root lignin concentration and supported via increased Oribatida abundance and decreased root K concentration. In contrast, legume presence showed a net positive effect on root mass loss via decreased root lignin concentration and increased root magnesium concentration, both of which led to enhanced root mass loss. Overall, the different measures of plant diversity had contrasting effects on root decomposition. Furthermore, we found that root chemistry and soil biota but not root morphology or soil abiotic conditions mediated these effects of plant diversity on root decomposition.